Corma raised $125 million at a reported $1 billion valuation, three months after emerging from stealth with AI models for cyber defense.
Cybersecurity startup Corma says it has raised $125 million in a Series A round at a $1 billion valuation. The Israeli company announced the financing November 16. That came about three months after it emerged from stealth with $60 million in seed funding. NEA led the round, alongside existing investors Sequoia Capital, Khosla Ventures and Coatue. Corma says the deal brings its total funding to $185 million and will accelerate its defensive cybersecurity work.
Corma’s $125 million Series A funding
The latest round followed Corma’s $60 million seed announcement in August. In roughly three months, the startup secured another $125 million. TechCrunch, Calcalist, Globes and The Wall Street Journal reported the financing and the company’s stated $1 billion valuation. That figure comes from a private funding round, not a public stock-market price.
NEA led the Series A, while three existing investors also participated. Sequoia Capital led Corma’s earlier seed round, which also included Khosla Ventures and Coatue. The new funding brings the startup’s total to $185 million, according to the company. Corma was founded in 2025, so its rapid fundraising followed its public launch, not its founding.
How Corma’s defensive cybersecurity AI works
Corma is developing foundation models trained for cyber defense and agents that can perform security work. Foundation models are AI systems that can be adapted for different tasks. Corma says its AI workforce works alongside security teams and uses their existing tools. The startup presents this as a specialized alternative to general-purpose AI.
CEO Alon Pluda told Calcalist that Corma is training models for cybersecurity. “We are training foundation models specifically for cybersecurity, models that are an order of magnitude better.” — Alon Pluda, Corma co-founder and CEO. That comparison is Pluda’s claim, not an independently established benchmark. The funding announcement provides no public test results to support it.
Corma says its agents can perform complex defensive tasks from start to finish. Security work can include reviewing alerts, logs and other system data to investigate possible threats. The company says its models are built for this workload and connect with tools already used by customers. Public materials do not explain the level of human oversight for every task.
Why investors are backing AI cyber defense
Corma’s funding comes as organizations face a growing cybersecurity workload. TechCrunch cited several software-security incidents disclosed on October 21, including vulnerabilities involving PHP and Node.js. LiteLLM also disclosed a security incident that day. These events illustrate the volume of issues security teams must manage, but do not prove AI caused each one.
AI may change how quickly attackers and defenders find weaknesses. Corma argues that models trained for coding do not necessarily meet enterprise defense needs. Security teams must interpret activity across systems and distinguish meaningful alerts from routine noise. Whether specialized models can improve this work at scale remains an open question.
Corma’s customer and adoption claims
Corma says large companies have adopted its models and delegated tasks to its AI workforce. Calcalist and Globes reported company claims of interest from Fortune 100, Fortune 500 and Fortune Global 500 organizations. The funding announcement did not name individual customers. It also gave no independently audited results on response times or attacks prevented.
This reported adoption helps explain investor interest, but public evidence remains limited. News outlets confirm the financing, while customer details largely come from Corma. The company says demand has exceeded its available computing capacity. That is the startup’s account, not an independent measure of market demand.
What Corma plans to do with the new funding
Corma says it will use the capital to expand model development and computing capacity. It also plans to add staff selectively. More computing power can support training and running larger AI systems, but does not guarantee stronger defenses. The company must show that its agents work reliably in complex security operations.
The rapid fundraising points to investor appetite for specialized cybersecurity AI. Still, a high private valuation cannot establish whether the technology will deliver lasting results. Customers will need to assess its performance, safety and cost as deployments expand. Those results can show whether Corma’s approach scales beyond early users.
Ultimately, Corma raised $125 million to accelerate its defensive cybersecurity AI work. The company says the round values it at $1 billion and brings total funding to $185 million. Its next challenge is proving that its models and agents help security teams. Wider deployments will offer a more meaningful test than the headline valuation.








